The mechanical behavior of the spinal column plays a key role in the overall structural response of the human skeleton. Since the stress and strain fields throughout the spinal column are not amenable to direct experimental measurements, the development of an analytical model is of practical importance for the study of normal and pathological spinal mechanics. This proposal requests computing time to continue our development of a three dimensional finite element model of a spinal column to study the mechanism of idiopathic scoliosis and to design better instrumentation for correction of the deformities of idiopathic scoliosis. This is a multi-year joint project between the Department of Biomedical Engineering and the Department of Orthopaedic Surgery at the University of Virginia. A finite element model (648 elements) based on CT scans of a 41-year-old woman's lumbar motion segment was developed for this project. A grant of 45 units in Cray Y-MP (CES930000N, NCSA, U. of Illinois at Urbana-Champaign) was utilized to develop the spinal motion segment model and the behavior of the motion segment was studied. Since one motion segment is not enough to study the behavior of a scoliotic spine, it is important to introduce a more realistic model including the entire spinal column for further studies. Nonlinear moving surface contact elements with iterative approach will be used to predict the behavior of the faced joints. The supercomputing system available at the National Center for Supercomputing at the University of Pittsburgh is best suited for studying this project.